Calc Electric Tank Standby Losses

[mo1258]

Does anyone know how to calculate the standby losses on an electric tank water heater?

I would think manufacturers would publish this data but I can't find it anywhere.

Here's one suggested estimate but it seems to yield very low costs so I am skeptical that it is accurate:
Assume indoor tank with 72 degree ambient room temp, no flows in or out
.93 Energy factor = 236K WH per year @ $0.10 = $24 per year
.94 Energy factor = 189 KWH per year @ $0.10 = $19 per year

Your thoughts? Thanks!

 

[Nukeman]

That looks about right. You could get a ballpark number by taking what the tag says that you would typically use in a year and mulitply by (1-energy factor). Really, the total has the energy factor already included (so a portion of it would be inculded twice with this basic calc), but it would get you a rough number.

Another way to do it would be to look at the annual operating cost on the tag and mulitply by (1-energy factor). For instance, a 50-gal Rheem Imperial is listed at 0.95 EF and has an annual operating cost of $492 (assuming 10.65 cents/KWH). $492*0.05 = $24.60/year

There will be additional standby losses from piping cooling down between uses or if the WH is located in an unconditioned space. However, the losses in the pipes will be the same for any type of WH used (insulate the pipes that you can get to).

Also, if the WH is in a conditioned space, the EF becomes 1.00 during the heating season as any of the "wasted" heat goes into warming the house, so the actual loss from the tank will actually be less than the $24 or so, especially if you live somewhere with a long heating season.

Electric units do very well in terms of standby loss as they have a heavy insulated blanket arond them. Gas models tend to have lower EF as the insulation is usually thinner as well as what is lost up the flue. However, gas may still be cheaper to run for a lot of people and gas also has a higher recovery rate. For my location, electricty is low and gas is somewhat high. The cost per BTU/KWH would be similar, but since I don't have other gas appliances, gas would cost me more as there is about a $13/month service charge for gas and that would kill any savings that I might get from a gas WH.

If you run the numbers, you'll find it very difficult to justify a tankless unit based on what is saved on standby losses (even if you were starting with a low EF gas unit). People also tend to use more hot water when they have a tankless since they have an "unlimited" supply. Tankless units are good if you don't have much space or have a large tub to fill (or a shower that is more like a car wash than a shower ).

 

[Sponsor]

 

[Jimbo]

I don't know if it is worth losing sleep over the amount. It is very small. I know when I used to turn apartments, the el cheapo gas water heater, probably .57 EF, would still have water @ >110º a week after the gas was off.

 

[Chris8796]

I would guess true standby tank loss would be very low. I would guess most losses occur through the piping. The exterior of a modern tank is at ambient temperature to the touch.

 

[Redwood]

Yep! Todays water heaters have such good insulation that calculating standby loss really isn't worth the bother...

Just like doing an insulation blanket isn't worth it.

Of course gas units with an undampered flue running up the middle do have room for improvement but electric is about as good as it gets.

 

[Dana]

Yep! Todays water heaters have such good insulation that calculating standby loss really isn't worth the bother...

Just like doing an insulation blanket isn't worth it.

Of course gas units with an undampered flue running up the middle do have room for improvement but electric is about as good as it gets.

Depends on your electricity cost, storage temp, and ambient though eh? In a 20cents/kwh utility, with 140F storage temp, installed in a garage with a 50F average ambient during the winter months the standby could easily justify the cost of the extra blanket. (And these installations DO exist...) 3 months of 90F delta-T at 2x the energy price is getting close to a year's worth of standby cost relative to 120-125F storage inside of fully conditioned space at the nat'l average electricity price, with another 9 months of lower-cost standby to go...

 

[mo1258]

Thanks for your replies to my post about Standby Losses. It is striking to me how "cheap" the losses are with a high efficiency electric.

 

[Ginahoy]

I have my electric water heater submetered, so measuring actual standby losses is easy. I did a couple of tests last spring** when I went out of town for a day or two and my water heater's standby losses are about 1.1 kw per day (about 150 Btu/hr) at 120F. This is consistent with theoretical heat loss equations.

** my heater is in the garage, so I figure testing in the spring the serves as a reasonable proxy for annual average.

Note that when calculating tank surface area and average r-value, keep in mind that the tank has rounded ends, and thus has less insulation (none at the bearing point). Also remember to subtract insulation thickness when estimating the tank dimensions. Unless you can find a tank schematic drawing, you'll have to guesstimate the radius ratio at the ends. Also remember that the benefit of additional insulation reduces as R-value increases (diminishing returns).

In my case, adding a 2" blanket (R6.7) might save about 10 kwh per month, or about $15 per year. So a $20 blanket would pay for itself in a little over a year (ignoring my labor). That's not a bad payback, though much longer than often touted.

 

[Thatguy]

If you're comfortable with electricity, measure the loss with an analog clock wired between one hot element terminal and ground, but first check that you actually do get approx. 120 V between these two terminals when the t'stat asks for heat.

If the clock advances 5 minutes in 7 hours overnight with no water use and a 4500 W element then the loss is 4500 [5/(7 x 60)] = 54 W. Before I added the blanket it was 5 minutes every 5 hours. This was done circa 1970 so the insulation may be better now.

When I did it, I used resistors to drop the 240 V to 120 V for the clock but sizing the resistors was tricky.

If you don't understand this post, don't do it this way.

 

[hj]

The bottom of the tank is "concave" so it does have insulation there. Although the top is convex, it still has a layer of insulation over it also.

 

[Ballvalve]

If you put all of your old goose down jackets and winter clothes over the heater, and old long underwear wrapped around any visible pipes, I would guess that 1.1 KW loss could go to .1 loss.

Down comforters work even better.

Stick with electric heaters however.

 

[Hackney plumbing]

If you're comfortable with electricity, measure the loss with an analog clock wired between one hot element terminal and ground, but first check that you actually do get approx. 120 V between these two terminals when the t'stat asks for heat.

If the clock advances 5 minutes in 7 hours overnight with no water use and a 4500 W element then the loss is 4500 [5/(7 x 60)] = 54 W. Before I added the blanket it was 5 minutes every 5 hours. This was done circa 1970 so the insulation may be better now.

When I did it, I used resistors to drop the 240 V to 120 V for the clock but sizing the resistors was tricky.

If you don't understand this post, don't do it this way.

Thats a very good idea. I like it.

 

[Runs with Bison]

Standby losses for electric are easier to calculate than for gas. The easiest way is to use the Energy guide usage and its basis to back out losses to the ambient. (Back calculate the kwh from the $ and cost per kwh printed on the sticker.) One would need to take a look at the test conditions required in the Federal Register. As I recall this was something like a 0.99 factor for resistance heating elements themselves, the remainder would be losses to ambient. I do not recall what the base ambient and water heater temps are for the test. Once that is found the effect of increasing the R-value with additional insulation (such as only on the walls) can be determined. This incremental savings can then be used to determine if the $/year saved justifies the insulation.

Another approach is to take the actual cylindrical tank dimensions from specs (walls and head) to calculate heat transfer area, then apply the foam insulation thickness for an R-value. From there use tank setpoint and ambient temps where the water heater is located to calculate heat loss. This is a very good measure of losses from the walls. The one thing it is missing is nozzles/projections: primarily the T&P valve and the inlet/outlet water lines. My understanding is that these would be included in the Energy Factor test/calculation protocol since it doesn't subtract them out from what I recall.